Permacomputing

Marloes de Valk
Ville-Matias Heikkila
Permacomputing

Shintaro Miyazaki: So how would you describe or define permacomputing?

Marloes de Valk: The term was coined and described first by Ville-
Matias in 2020 on his website? and has since started circulating, being
connected to diverse practices. Currently it is collaboratively
developed on the permacomputing wiki.* I’m very interested in how the
term’s meaning will develop now that people are adopting it. The term
resonates with a lot of people, it captures something that is very much
wished for, a counternarrative to the rapid “upgrade-or-die” cycle
promoted by the tech industry. It is also somewhat paradoxical, linking
an environmentally lightweight practice such as permaculture with one
that is as resource-intensive as computing, which doesn’t only weigh
heavy on the planet due to energy consumption, but also through
unethical and damaging practices in several parts of the supply chain,
from mining to manufacturing. Most energy is used during the production
and spectacularly wasteful end-of-life phase. This is unequally
affecting the Global South. The term permacomputing hints at wanting to
do better without being naive about this paradox.

Ville-Matias Heikkilä: Permacomputing asks the question whether we can
rethink computing in the same way as permaculture rethinks agriculture.
Is there even place for high technology (such as computing) in a world
where human civilizations contribute to the well-being of the biosphere
rather than destroy it? Permacomputing wants to imagine such a place
and take steps towards it. It is therefore both utopian and practical.
From today’s viewpoint it can indeed also feel paradoxical, because
current mainstream computing is pretty much the type of technology that
epitomizes industrial wastefulness and many other things that have gone
wrong in society and technology.

"Perma” refers to permanence. A technology that depends on a wasteful
use of finite resources can hardly be permanent. This is why a radical
reduction of that wastefulness is a major concern to us: maximize the
hardware lifespans, minimize the energy use. And this is not just about
a set of technical problems to be fixed - the attitudes also need a
radical turn. Small is beautiful, understandability is beautiful,
"virtual” is not immaterial, online time should be used wisely, not
everything needs to be constantly available, doing things with less is

N-AALNNOD

not "returning to the past.” For a more thorough explanation of the
ideas, see my web articles on the topic (the 2020 original and the 2021
update).3

As for now, permacomputing is more like a set of interrelated ideas and
practices than a coherent system of thought. Different people have
different views on it. Also, many people have independently come up
with similar ideas, so there’s a lot of overlap between permacomputing
and concepts such as "salvage computing,” “frugal computing,” and
"degrowth computing.” Much of what comes out of the annual Computing
within Limits workshops also has a lot of overlap with permacomputing.
I think we should regard all these ideas as different aspects of the
same thing, rather than as separate movements whose borders need to be
unambiguously defined.

Marloes de Valk: Connecting these diverse yet overlapping practices is
very important because they are often not visible to a general
audience, sometimes not even to each other. The idea that there are no
alternatives to mainstream tech platforms needs to be challenged, the
greenwashing on their sustainability pages needs to be exposed. There
is a wealth of practices in existence, historical and current, that
approach computing as needing limits, including permacomputing, and
making the different interlinked ideas visible, showing the relations
between them, and creates a strong counternarrative to the one of green
capitalism,* which is shaping public discourse on how to best deal with
the climate crisis, yet doesn’t challenge the “business as usual” of
free market capitalism that caused the crisis in the first place. The
different terms and related practices that counter this are not
competing for recognition as “the” best solution, they are
strengthening each other. My hope is that communities working on
developing, repairing, and maintaining their own computing
infrastructure will benefit from each other’s experiences, while always
applying ideas and technologies in a situated way, adapting them to
meet their local and specific needs.

Shintaro Miyazaki: How is permacomputing different from let’s say
supercomputing or feminist computing?

Marloes de Valk: Ville-Matias worded this nicely, in the aforementioned
essay: “The existence of computers in a human civilization can be
justified by their ability to augment the potential”® for humans to have
a strengthening effect on the ecosystem. Supercomputing can be
essential to accomplish this, for instance when calculating climate
models, yet this is only one application of supercomputing. I consider
it a technology, not a philosophy. It can be used for climate research
but also for something destructive such as oil and gas exploration. It
was used to model the spread of the coronavirus, to forecast extreme
weather, and to simulate nuclear tests. One of the main goals of

permacomputing is to use as little artificial energy as possible, in
that respect supercomputing should be used extremely sparingly because
it consumes massive amounts of energy; both in the form of electricity
and as embodied energy in the hardware.

Ville-Matias Heikkila: I don’t think supercomputing is a specific
technology but a design philosophy that aims at maximizing computer
performance at any cost. There are some scientific problems where
results actually improve the more you scale the model (weather
forecasting for example), but supercomputing has also been a major
guiding principle of mainstream computing for decades (e.g., the holy
grail of photorealism in consumer computer graphics), and people even
find it difficult to think outside the “more is better” box.
Permacomputing is all about a wise use of resources. Huge computing
tasks (especially ones that help heal the biosphere etc.) are something
a society can use some of its surplus energy for, but our main focus is
in “less is more” and the minimization of the use of artificial energy.
Being happy with the abundance we already have instead of regarding it
as scarcity that needs to be overcome with premature obsolescence.

‘ST TYAXTOH SETIEW-STITA PUL ATeA ep SƏOTILEN

One can say that the difference between permacomputing and “mainstream
computing” is “by definition” the same as the difference between
permaculture and industrial agriculture. In industrial agriculture, you
pour more and more artificial energy and resources on the land in order
to maximize the production in a simple-minded, straightforward, and
“scalable” way, while in permaculture you co-operate with the land in
inventive and thoughtful ways that get you a fair yield with minimal
use of artificial energy. The idea of supercomputing crystallizes quite
well the mindset that permacomputing is offering an alternative to.

butqndwovewrəd

Mar loes de Valk: There are many overlaps between what I understand to
be feminist computing (there is not one single definition or practice
exemplifying it, but many) and permacomputing.

Ville-Matias Heikkilä: It seems to me that movements such as feminist
computing often have their main focus on social justice issues around
the technology, while permacomputing is mainly focused on the why’s and
how’s of the technology itself. In the 19th century, some activists
raised issues about the exploitation that took place in factories,
while some others entirely questioned the maximalist production model
that factories represented. Both were right and equally important.

Marloes de Valk: I think that is the strength of an intersectional
feminist perspective on technology, a view in which technology is never
disconnected from social reality, lived and embodied experience.
Exploitation and a maximalist production model are inseparable, you
cannot maximalize without exploiting people and planet. The differences

N-AALNNANOD

between permacomputing and feminist computing are perhaps not as
interesting as thinking about how these terms intersect.

For instance the refusal to apologize for not always being available as
worded in the Feminist Server Manifesto (FSM)® is not only related to
the labor needed to secure maximum uptime, it is also leading to a
rethinking of availability in terms of ecological impact. To accomplish
5 nines uptime, a 99.999% availability of servers, the infrastructure
needs to expand enormously; for every extra nine a massive amount of
artificial energy is required in the shape of hardware (representing
embodied energy and resource use), electricity, and fossil fuels.
Datacenters burn through millions of liters of diesel every year just
to test their emergency generators for example.

Another interesting overlap is the Feminist Server Manifesto’s
statement to “not strive for seamlessness. Talk of transparency too
often signals that something is being made invisible.” In
permacomputing and permaculture alike, observation is important. You
captured this well when writing about how current consumer-oriented
computing systems go to great lengths to hide the inner workings of a
system, alienating people from their devices. You suggest computer
systems could make their inner workings as observable as possible.
There is another aspect of observation that deserves attention. People
are very much craving stories on how we can approach computing
differently and are eager to learn what can be done concretely and
practically. The hard part, it seems, is understanding that in terms of
sustainability, most of it comes down to not doing, not buying, not
adding, yet the initial response is often an impulse purchase of a
solar panel and a Raspberry Pi... In certain cases a valid investment,
but not always, and careful observation of a problem might lead to
different conclusions and solutions. This made me think of a section in
a book on permaculture that explained how the first year with a new
piece of land you do nothing but observing. Only after having seen the
land and all its inhabitants in the process of all seasons can you know
what could be done with minimal waste of energy for all species
involved, human and non-human.

Ville-Matias Heikkila: I might also include situatedness. It is
mentioned in the FSM as the first bullet point and is also highly
relevant to permacomputing. Computers and technology in general should
integrate to their surroundings both physically and culturally,
adapting to the changing energy conditions and to the needs of the
local people. It is important to decentralize and localize.

Systems like WWW that encourage excessive centralization have led us to
a kind of world where being able to access a distant centralized server
99.999% of the time is something to strive for and where the owner of a
huge server can enforce a “one size fits all” model to all of its users

regardless of their cultures and local needs. And this centralized,
“one size fits all” mentality also makes it difficult for people to
envision alternative computing cultures, and the lack of this kind of
imagination makes it easier for corporations to impose draconian
limitations on how people can use technology. Having a more situated,
decentralized, and diverse computer culture would help a lot of things.

I would also like to mention the importance of historical situatedness.
When permacomputing is envisioning computing several decades or
centuries into the future, it is important to know where we came from.
The mainstream history of computing is often a “winners’ history” told
from a Moore’s law point of view: ever-more powerful hardware
“generations” making their predecessors obsolete. We therefore
desperately need more varied and alternative approaches to the history
of computing in order to better situate ourselves in it. Local
histories are particularly important. In a lot of Europe, for example,
local computing cultures were “colonized” and even “deprecated” in the
1990s by a “Silicon Valley” culture that emphasized constant growth and
obsolescence. This event is sometimes called Siliconization. If we only
know about the “siliconized” idea of computing, it is very difficult to
envision non-siliconized futures. I’m personally a part of the
demoscene, a computer art subculture that has managed to keep many
“ore-Siliconization” ideals and practices: strict technological
constraints can amplify creativity, small is beautiful, no platform is
obsolete. Had there not been this kind of “alternative world” to
contrast mainstream computing against, coming up with the idea of
permacomputing might have taken much longer.

Shintaro Miyazaki: Where do you situate “countering” in permacomputing?
How would you situate permacomputing within countering and how would
you imagine its role?

Marloes de Valk: Countering within permacomputing is found in providing
a counternarrative to the idea of eternal growth and its counterpart
overproduction, through a “less is more” approach, exemplified by
working with (e-)waste as a resource, emphasizing repair and
maintenance rather than production of new devices, planned longevity
rather than planned obsolescence. Permacomputing also counters the idea
that computers “dematerialize” and lessen resource use, and does so
through resource-sensitivity: adapting energy use to energy
availability and using energy thoughtfully. Permacomputing counters
tech corporations’ claims to universality through a decentralized and
local approach. Decentralization is of course not by definition more
sustainable, as exemplified by cryptocurrencies and blockchain
technology. In the context of permacomputing, decentralization refers
to autonomously-administered servers and services, for and by
communities. This is also not by definition more sustainable, but at
least gives communities the power to make choices about their tools

‘STTYAATOH SETIEW-STITA PUL ATPA ep seo T7eW

Hutyznduooeuz38g

N-AALNNOD

that allow for less energy-hungry practices, such as offline first,
adaptability to intermittent connectivity, and running software that
uses as little energy as possible and is tailored to local needs.’

Ville-Matias Heikkila: Back in around 2008, I named my blog
“Countercomplex.” The main idea in that decision was that I wanted to
counter the kind of complexity that was prominent in the computing
world by embracing the smallest, the bit-level minimalism. On the other
hand, the countercomplex plane is a mathematical term for a plane that
is just as complex as the ordinary complex plane but in an alternative
way, so I could also use it as a metaphor for how complexity and
diversity may emerge from very small seeds. The Bytebeat phenomenon, or
how very short formulas of largely bitwise operations may yield
surprisingly-complex structured music, was a kind of prime example of
this type of “countercomplexity.” If even very short one-liner programs
can yield something like this, we have possibly only scratched the
surface of what even very small computers can do. This is also where I
envision the potential of “growth” in permacomputing: explore the
unexplored spaces of the devices we already have instead of getting
stuck with the ideas that their designers had. Computing is a huge
ocean of possibilities, even within strict limits, and that’s why we
should counter the misguided idea that new things only become possible
by “technological progress” that is synonymous with increased
complexity.

Shintaro Miyazaki: Please suggest further counter-Ns, N-computing(s),
or N-futuring(s). Or other scholars/artists to have a conversation
with.

Ville-Matias Heikkila: Computing within Limits in general is full of
interesting ideas we have found relevant to permacomputing and many
other ecological alternative-computing concepts, so the people involved
with it would be my first suggestion.

Marloes de Valk: I agree, the workshop takes a more radical and
transdisciplinary approach than other conferences and workshops on
sustainability and ICT. It brings together a growing group of scholars
dedicated to promoting the design of computing contributing to a
transition to a future in which the well-being of humans and other
species is the primary objective.® The workshop’s publications contain a
wealth of interesting terms and related design principles. The term
Computing within Limits brings together three principles, according to
a 2018 paper by Nardi et al.: it questions growth and aims instead for
a steady-state economy, it considers models of scarcity in order to
promote resilience in a diversity of current and future contexts, and
lastly it aims at reducing energy and material consumption while
avoiding the Jevons paradox or rebound effect, in which gains in
efficiency often result in lower costs, a subsequent growing demand,

and increased resource consumption. The workshop’s focus and the
interpretation of its title have shifted a little over the years,
starting with an emphasis on designing in the abundant present for the
use in a future of limits, to designing for the present with an
awareness of current real-world limits.

I’d like to suggest one more N-computing: Trans*feminist computing
(Cell for Digital Discomfort) because even though it speaks of
practices related to servers instead of computing in general, it offers
pluriform and non-cis imaginations with its careful and collective
expansion of the feminisms of the Feminist Server Manifesto as an
“ambiguous ongoing wishlist for techno-ecologies in the making; an
ongoing set of spells for a different tech for this world, for
different tech for different worlds.”1°

1 Ville-Matias Heikkilä, “Permacomputing,” Viznut, 24 June 2020, available at

http://viznut.fi/texts-en/permacomputing. html.

2 Permacomputing wiki, 2022, available at https://permacomputing.net.

3 Ville-Matias Heikkilä, “Permacomputing update 2021,” Viznut, 27 August 2021,
available at http://viznut.fi/texts-en/permacomputing_update_2021.html.

4 Marloes de Valk, “A pluriverse of local worlds: A review of Computing within
Limits related terminology and practices,” Computing within Limits, 14 June 2021,

available at https://limits.pubpub.org/pub/jkrofgqlk/.
5 Ville-Matias Heikkilä, “Permacomputing,” Viznut, 24 June 2020, available at

http://viznut.fi/texts-en/permacomputing. html.
6 “A Feminist Server Manifesto 0.01,” Are You Being Served? 2014, available at

https ://areyoubeingserved.constantvzw.org/Summit_afterlife. xhtml.
7 Decentralization is described in more detail on the permacomputing wiki:

https ://permacomputing.net/decentralization/.
8 Bonnie Nardi, Bill Tomlinson, Donald J. Patterson, Jay Chen, Daniel Pargman,

n

Barath Raghavan, and Birgit Penzenstadler, “Computing within Limits,” Communications
of the ACM, vol. 61, no 10. 86-93, available at https://doi.org/10.1145/3183582.
available at https://computingwithinlimits.org.
10 “Trans*feminist servers,” pamphlet circulated at Art Meets Radical Openness:

DEBUG, Linz, Austria, 15-18 June 2022.

7

9 See “Computing within Limits,”

7

‘ST TYAXTOH SETIEW-STITA PUL JTeA ep SƏOTILEN

butqndwovewrəd

N-AALNNANOD

Marloes de Valk and Ville-Matias Heikkilä
Permacomputing

This publication is part of COUNTER-N, an online journal edited

by Özgün Eylül İşcen and Shintaro Miyazaki. It is published electronically
on the edoc-server of the Humboldt-Universität zu Berlin,
https://edoc.hu-berlin.de/

Title: Permacomputing

Authors: Marloes de Valk and Ville-Matias Heikkilä
Editors: Özgün Eylül İşcen, Shintaro Miyazaki
Graphic Design: Ahmet M. Öğüt

Published in 2022 by counter-n.net

DOI: doi.org/10.18452/25609
Final Submission: July 15th, 2022

This Publication is licensed under the CC-BY-SA 4.0 International.
To view a copy of this license,
visit: http://creativecommons.org/licenses/by-sa/4.0/.